SEARCH · Engineering Papers

Results for “input estimation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,081 records · Page 60

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Modeling of gap conductance for LWR fuel rods applied in the BISON code

A new gap conductance model is proposed in this study as a combination of Toptan’s model and the Ross-Stoute model. A variance-based sensitivity analysis is performed to understand how simulation results depend on all input parameters of the proposed model. Additionally, new modeling options (e.g. fill gas thermal conductivity, temperature jump distance, thermal accommodation coefficient, etc.) are added into the nuclear fuel performance code, BISON. The need for further investigation of the gap heat transfer between fuel and cladding in BISON motivated this study to evaluate its impact on the code’s predictions. New gap conductance modeling is proposed. A series of integral-effects validation tests is performed: (1) to demonstrate the impact of the proposed model on the code’s fuel temperature predictions at the beginning of life and through the reactor’s life; (2) to ensure that the proposed model is capable of accurately modeling gap heat transfer characteristics in real-world problems; and (3) to investigate the impact of the estimation of fission gas release on the fuel temperature predictions with the proposed model. Overall, the results indicate that the proposed gap conductance model improves BISON’s predictions.

42 ENGINEERING↗

DOE OSTI2022

AN ESTIMATE OF SPENT NUCLEAR FUEL MECHANICAL LOADS IN THE GENERAL 30 CM PACKAGE DROP SCENARIO

The US Department of Energy Spent Fuel and Waste Science and Technology (SFWST) program is performing research to determine the mechanical loading conditions applied to spent nuclear fuel (SNF) during normal conditions of transport to inform mechanical tests of SNF and close an important knowledge gap related to the practical disposition of SNF in the US. Researchers at Pacific Northwest National Laboratory (PNNL) have completed an extensive finite element study to characterize and estimate the potential mechanical loads on SNF during a hypothetical 30 cm drop of an SNF transportation package. This modeling study is validated with test data collected by the SFWST program during a physical test campaign that included one-third scale package drop tests and full-scale single fuel assembly drop tests. The test campaign was led by Sandia National Laboratories (SNL) and included international collaboration with Equipos Nucleares S.A, S.M.E (ENSA) and Bundesanstalt für Materialforschung und -prüfung (BAM). The key contribution of the modeling study is to go beyond the limitations of the limited number of physical tests to estimate the impact response to variations in impact angle, initial gap conditions, fuel assembly design, burnup and other parameters that affect the mechanical loads. The methodology of this study included a classic parametric study to calculate the impact response of highly detailed fuel assemblies over many combinations of parameters. Models of a 17x17 pressurized water reactor fuel assembly and a generic 10x10 boiling water reactor fuel assembly were both used in this study to cover the major fuel assembly types in the US inventory. Over 2,000 impact responses were calculated. The results of the parametric study were evaluated using traditional methods and basic statistics. The results were also used to construct a damage model using multiple nonlinear regression techniques to predict the mechanical loads over the full range of all input parameters. The damage model was found to work very well for all impact angle cases where the cask came to rest on its side. It was concluded that end drop cases where the cask remained vertical (instead of tipping over onto its side) were not sufficiently characterized by the current set of parametric study cases to include in the damage model, but it was not a priority to fully investigate that range because the highest mechanical loads were observed in the broader range of side impact cases. This modeling work provides sufficient insight into the mechanical loads on SNF during a hypothetical 30 cm package drop that, when considered along with the physical test data collected by the SNL-led team, the SFSWT program can consider the knowledge gap closed.

Klymyshyn, Nicholas A.↗

DOE OSTI2022

Generative Adversarial Networks and Mixture Density Networks-Based Inverse Modeling for Microstructural Materials Design

Abstract There are two broad modeling paradigms in scientific applications: forward and inverse. While forward modeling estimates the observations based on known causes, inverse modeling attempts to infer the causes given the observations. Inverse problems are usually more critical as well as difficult in scientific applications as they seek to explore the causes that cannot be directly observed. Inverse problems are used extensively in various scientific fields, such as geophysics, health care and materials science. Exploring the relationships from properties to microstructures is one of the inverse problems in material science. It is challenging to solve the microstructure discovery inverse problem, because it usually needs to learn a one-to-many nonlinear mapping. Given a target property, there are multiple different microstructures that exhibit the target property, and their discovery also requires significant computing time. Further, microstructure discovery becomes even more difficult because the dimension of properties (input) is much lower than that of microstructures (output). In this work, we propose a framework consisting of generative adversarial networks and mixture density networks for inverse modeling of structure–property linkages in materials, i.e., microstructure discovery for a given property. The results demonstrate that compared to baseline methods, the proposed framework can overcome the above-mentioned challenges and discover multiple promising solutions in an efficient manner.

36 MATERIALS SCIENCE↗

DOE OSTI2026

Self-supervised and multi-fidelity learning for extended predictive soil spectroscopy

Infrared spectroscopy is a cost-effective, non-destructive, and environmentally benign technology that is increasingly recognized as an important solution for meeting the global demand for soil data. While both near-infrared (NIR) and mid-infrared (MIR) diffuse reflectance spectroscopy enable rapid estimation of soil properties, they present a significant trade-off: NIR offers superior scalability and lower operational costs, whereas MIR provides higher analytical fidelity by capturing fundamental molecular vibrations. In this study, we propose a self-supervised, multi-fidelity learning framework designed to bridge this gap. Our approach leverages large-scale MIR spectral libraries to learn a compact, transferable latent representation, into which NIR spectra are subsequently aligned for downstream prediction. The workflow consists of pretraining a latent model on a large MIR library, adapting the representation using a smaller paired NIR–MIR dataset, and evaluating generalization on an independent external test set. Across a range of chemical and physical soil properties, we found that MIR-derived embeddings improved prediction accuracy relative to baseline models that used raw MIR inputs. Predictions derived from the spectrum conversion (NIR to MIR) task did not match the performance of the original MIR spectra but were similar or superior to predictive performance of NIR-only models, suggesting the unified spectral latent space can effectively leverage the larger and more diverse MIR dataset for prediction of soil properties not well represented in current NIR libraries.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2020

Quantifying heavy quark transport coefficients with an improved transport model

We report that the heavy-flavor transport coefficients contain important information on the strong interaction at finite temperatures. The extraction of these numbers from experimental data requires dynamical modeling of heavy-flavor transport that is coupled to realistic medium evolution. Furthermore, meaningful extractions necessitate both a faithful implementation of the physical inputs to be tested and the quantification of model uncertainty. For these purposes, we have developed a partonic transport model LIDO. It has an improved treatment of in-medium parton bremsstrahlung, which has been calibrated to theoretical calculations in a simple medium to reduce modeling uncertainty. Regarding the interaction between heavy quark and the medium, few-body perturbative scatterings are applied to large-momentum transfer (q) processes, while a diffusion equation models the dynamics of small-q processes. Such a separation restricts the explicit use of medium quasi-particles to large-q processes only. Another advantage is that deviations from the leading-order probe-medium coupling can be parametrized as an additional contribution to the diffusion constant. The heavy quark transport coefficients are then extracted with uncertainty estimation from a Bayesian analysis including both the RHIC and the LHC data. The results are found to be consistent with earlier extraction of the light-quark transport coefficients at high momentum and be comparable with lattice calculations of the heavy-flavor diffusion constant in the static limit at low momentum.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2022

Machine learning enabled quantification of the hydrogen bonds inside the polyelectrolyte brush layer probed using all-atom molecular dynamics simulations

The configuration of densely grafted charged polyelectrolyte (PE) brushes is strongly dictated by the properties and behavior of the counterions that screen the PE brush charges and the solvent molecules (typically water) that solvate the brush molecules and these screening counterions. Only recently, efforts have been made to study the PE brushes atomistically, thereby shedding light on the properties of brush-supported ions and water molecules. However, even for such efforts, there are limitations associated with using a generic definition to estimate certain properties of water and ions inside the brush layer. For example, water–water hydrogen bonds (HBs) will behave differently for locations outside and inside the brush layer, given the fact that the densely closely grafted PE brush molecules create a soft nanoconfinement where the water connectivity becomes highly disrupted: therefore, using the same definition to quantify the HBs inside and outside the brush layer will be unwise. In this paper, we address this limitation by employing an unsupervised machine learning (ML) approach to predict the water–water hydrogen bonding inside a cationic PE brush layer modeled using all-atom molecular dynamics (MD) simulations. Here, the ML method, which relies on a clustering approach and uses the equilibrium coordinates of the water molecules (obtained from the all-atom MD simulations) as the input, is capable of identifying the structural modification of water–water HBs (revealed through appropriate clustering of the data) inside the PE brush layer induced soft nanoconfinement. Such capabilities would not have been possible by using a generic definition of the HBs. Our calculations lead to four key findings: (1) the clusters formed inside and outside the brush layer are structurally similar; (2) the margin of the cluster is shorter inside the PE brush layer confirming the possible disruption of the HBs inside the PE brush layer; (3) the average “hydrogen–acceptor-oxygen–donor-oxygen” angle that defines the HB is reduced for the HBs formed inside the brush layer; (4) the use of the generic definition (definition usable for characterizing the HBs in brush-free bulk) leads to an overprediction of the number of HBs formed inside the PE brush layer.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

The NuMI Flux Prediction at ICARUS

The Deep Underground Neutrino Experiment (DUNE) is a next-generation long-baseline neutrino oscillation experiment seeking to probe fundamental symmetries within the structure of the Pontecorvo-Maki-Nakagawa-Sakata (PMNS) mixing matrix, and perform precision measurements its parameters including the neutrino mass ordering via the sign of $\Delta m^2_{31}$, and the charge-parity violating phase, $\delta_{CP}$. To make these measurements with high precision, DUNE will require external $\nu$-Ar scattering cross section data as a crucial input to the oscillation fit. Imaging Cosmic And Rare Underground Signals (ICARUS) is a 476 t liquid argon neutrino detector located at Fermi National Accelerator Laboratory (FNAL) where it is serving as the far detector for the Short-Baseline Neutrino (SBN) program along the Booster Neutrino Beam (BNB) axis. ICARUS additionally lies 795 m downstream and 100.1 mrad off-axis of the Neutrinos at the Main Injector (NuMI) neutrino beam. From this position, ICARUS is exposed to a large flux of NuMI (anti-)electron and (anti-)muon neutrinos, and poses a unique opportunity to provide high-statistics measurements of quasi-elastic and single pion-production cross sections for four neutrino flavors ($\nu_{\mu}$, $\nu_{e}$, $\bar{\nu}_{\mu}$, $\bar{\nu}_{e}$). This dissertation is centered around accurately characterizing the models and estimating their precision for use in making these measurements. This includes identifying major sources of uncertainty in the models such that they can be properly propagated to the cross section measurements. Specifically, this work focused on the model of the NuMI beamline and its impact on the neutrino fluxes, but also delved into the detector response model and its impact on reconstructed observables in the detector. Significant efforts were made to improve the characterization to enhance precision and thus reduce the level of propagated uncertainty. In particular, the NuMI flux was determined to be composed of 57% $\nu_{\mu}$, 38% $\bar{\nu}_{\mu}$, 3% $\nu_{e}$, and 2% $\bar{\nu}_{e}$ while the horns are operating in the positive-particle focusing configuration. The total uncertainty on the $\nu_{\mu} + \bar{\nu}_{\mu}$ ($\nu_{e} + \bar{\nu}_{e}$) flux while operating in the forward horn operating mode was determined to be 10.84% (9.04%). Compared to the on-axis flux, mesons that eventually decay to neutrinos more frequently reinteract within the NuMI structure, resulting in elevated uncertainty as these processes are not well-constrained by existing hadron interaction cross section measurements. Covariance matrices were calculated to propagate the flux uncertainty characterization to NuMI analyses.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2024

Denoising Seismic Waveforms Using a Wavelet-Transform-Based Machine-Learning Method

Seismic waveform data recorded at stations can be thought of as a superposition of the signal from a source of interest and noise from other sources. Frequency‐based filtering methods for waveform denoising do not result in desired outcomes when the targeted signal and noise occupy similar frequency bands. Recently, denoising techniques based on deep‐learning convolutional neural networks (CNNs), in which a recorded waveform is decomposed into signal and noise components, have led to improved results. These CNN methods, which use short‐time Fourier transform representations of the time series, provide signal and noise masks for the input waveform. These masks are used to create denoised signal and designaled noise waveforms, respectively. However, advancements in the field of image denoising have shown the benefits of incorporating discrete wavelet transforms (DWTs) into CNN architectures to create multilevel wavelet CNN (MWCNN) models. The MWCNN model preserves the details of the input due to the good time–frequency localization of the DWT. In this report we use a data set of over 382,000 constructed seismograms recorded by the University of Utah Seismograph Stations network to compare the performance of CNN and MWCNN‐based denoising models. Evaluation of both models on constructed test data shows that the MWCNN model outperforms the CNN model in the ability to recover the ground‐truth signal component in terms of both waveform similarity and preservation of amplitude information. Model evaluation of real‐world data shows that both the CNN and MWCNN models outperform standard band‐pass filtering (BPF; average improvement in signal‐to‐noise ratio of 9.6 and 19.7 dB, respectively, with respect to BPF). Evaluation of continuous data suggests the MWCNN denoiser can improve both signal detection capabilities and phase arrival time estimates.

58 GEOSCIENCES↗

DOE OSTI2022

Fast decay of classification error in variational quantum circuits

Variational quantum circuits (VQCs) have shown great potential in near-term applications. However, the discriminative power of a VQC, in connection to its circuit architecture and depth, is not understood. To unleash the genuine discriminative power of a VQC, we propose a VQC system with the optimal classical post-processing—maximum-likelihood estimation on measuring all VQC output qubits. Via extensive numerical simulations, we find that the error of VQC quantum data classification typically decays exponentially with the circuit depth, when the VQC architecture is extensive—the number of gates does not shrink with the circuit depth. This fast error suppression ends at the saturation towards the ultimate Helstrom limit of quantum state discrimination. On the other hand, non-extensive VQCs such as quantum convolutional neural networks are sub-optimal and fail to achieve the Helstrom limit, demonstrating a trade-off between ansatz complexity and classification performance in general. To achieve the best performance for a given VQC, the optimal classical post-processing is crucial even for a binary classification problem. To simplify VQCs for near-term implementations, we find that utilizing the symmetry of the input properly can improve the performance, while oversimplification can lead to degradation.

Zhang, Bingzhi↗

DOE OSTI2020

Leveraging operational information from wastewater injection wells to evaluate CO 2 injection performance for carbon storage applications in the Appalachian Basin

Abstract Geologic parameters, geophysical logging, injection testing, and operational metrics from wastewater injection wells were integrated to develop a preliminary design of a carbon storage facility in the Appalachian Basin. A scattered group of 10–20 commercial wastewater injection wells dispose off produced water from oil and gas wells in the region, utilizing a sequence of stacked deep saline formations for injection zones. These wastewater injection wells provide practical benchmarks for understanding the feasibility of carbon dioxide (CO 2 ) storage. Geologic models were developed based on characterization data from the wastewater injection wells. Reservoir simulations were calibrated according to injection testing and operational data from the wastewater injection wells. Long‐term operational data on injection flow rates and pressures measured in the wastewater injection wells were especially useful to evaluate the performance of carbon storage applications. The simulations were used to estimate injection pressures, radius of CO 2 saturation, and pressure response for industrial scale CO 2 storage applications. Results were also used to provide a design basis in terms of number of injection wells, well spacing, area of review, injection system components, monitoring plan, and CO 2 pipeline distribution system. The analysis demonstrates that there is sufficient injectivity in the deep saline formations in the west‐central Appalachian Basin to store commercial volumes of anthropogenic CO 2 . The geologic system appears suitable for supporting CO 2 injection rates of 0.5–1.0 million metric tons per year at injection pressures below formation fracture pressure in a single well. The long‐term operational data of wastewater injection wells within the study area suggested a lower permeability‐thickness values than indicated by initial reservoir tests. A workflow for developing realistic permeability values for input into reservoir simulations is presented. © 2020 Society of Chemical Industry and John Wiley & Sons, Ltd.

Valluri, Manoj↗

DOE OSTI2025

Introducing the SLICE Method for estimating pebble-bed reactor inventories at equilibrium operation with SCALE

This paper introduces the SCALE Leap-In method for Cores at Equilibrium (SLICE) for estimating pebble-bed reactor equilibrium core isotopic inventories using capabilities in the SCALE code system, requiring only a small computational cluster and a few days of computation. This method uses an iterative approach that relies on (1) a surrogate spectrum model that captures spatial and time-dependent spectral conditions, (2) a multi-pass model that captures the pebble’s evolving nuclide inventory as a function of location and time in the core, and (3) a full-core model that captures the core’s spatial neutron flux distribution. The SLICE approach is applied to a generic fluoride salt–cooled high-temperature reactor, demonstrating fuel inventory convergence through nuclide concentration inspection across iterations and comparisons for core realizations with varying discretizations. Results agree within ~5% with another state-of-the-art code, with differences attributed to input parameter or modeling assumption variations in the equilibrium generation methods.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2021

Active microbial biomass decreases, but microbial growth potential remains similar across soil depth profiles under deeply-vs. shallow-rooted plants

Climate-smart land management practices that replace shallow-rooted annual crop systems with deeply-rooted perennial plants can contribute to soil carbon sequestration. However, deep soil carbon accrual may be influenced by active microbial biomass and their capacity to assimilate fresh carbon at depth. Incorporating active microbial biomass, dormancy, and growth in microbially-explicit models can improve our ability to predict soil's capacity to store carbon. But, so far, the microbial parameters that are needed for such modeling are poorly constrained, especially in deep soil layers. Here, we used a lab incubation experiment and growth kinetics model to estimate how microbial parameters vary along 240 cm of soil depth in profiles under shallow- (soy) and deeply-rooted (switchgrass) plants 11 years after plant cover conversion. We also assessed resource origin and availability (total organic carbon, 14 C, extractable organic carbon, specific UV absorbance of K 2 SO 4 extractable organic C, total nitrogen, total dissolved nitrogen) along the soil profiles to examine associations between soil chemical and biological parameters. Even though root biomass was greater and rooting depth was deeper under switchgrass than soy, resource availability and microbial growth parameters were generally similar between vegetation types. Instead, depth significantly influenced soil chemical and biological parameters. For example, resource availability and total and relative active microbial biomass decreased with soil depth. Decreases in the relative active microbial biomass coincided with increased lag time (response time to external carbon inputs) along the soil profiles. Even at a depth of 210–240 cm, microbial communities were activated to grow by added resources within a day. Maximum specific growth rate decreased to a depth of 90 cm and then remained consistent in deeper layers. Our findings show that >10 years of vegetation and rooting depth changes may not be long enough to alter microbial growth parameters, and suggest that at least a portion of the microbial community in deep soils can grow rapidly in response to added resources. Our study determined microbial growth parameters that can be used in microbially-explicit models to simulate carbon dynamics in deep soil layers.

14C↗

DOE OSTI2020

Historical total ozone radiative forcing derived from CMIP6 simulations

Radiative forcing (RF) time series for total ozone from 1850 up to the present day are calculated based on historical simulations of ozone from 10 climate models contributing to the Coupled Model Intercomparison Project Phase 6 (CMIP6). In addition, RF is calculated for ozone fields prepared as an input for CMIP6 models without chemistry schemes and from a chemical transport model simulation. A radiative kernel for ozone is constructed and used to derive the RF. The ozone RF in 2010 (2005–2014) relative to 1850 is 0.35 W m -2 [0.08–0.61] (5–95% uncertainty range) based on models with both tropospheric and stratospheric chemistry. One of these models has a negative present-day total ozone RF. Excluding this model, the present-day ozone RF increases to 0.39 W m -2 [0.27–0.51] (5–95% uncertainty range). The rest of the models have RF close to or stronger than the RF time series assessed by the Intergovernmental Panel on Climate Change in the fifth assessment report with the primary driver likely being the new precursor emissions used in CMIP6. The rapid adjustments beyond stratospheric temperature are estimated to be weak and thus the RF is a good measure of effective radiative forcing.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Castanha et al. 2018: Hopland Lysimeter 13C-Labeled Root Litter Decomposition Study.

The breakdown and decomposition of plant inputs are critical for nutrient cycling, soil development, and climate-ecosystem feedbacks, but uncertainties persist in how the rates and products of litter decomposition are affected by soil temperature, rhizosphere, and depth of input. To elucidate these questions we measured the effects of soil warming (+ 4 °C), rhizosphere, and depth of litter placement on the decomposition of Avena fatua (wild oat grass) root litter in a Mediterranean grassland ecosystem. Field lysimeters were subjected to three environmental treatments (heating, control, and plant removal) and three 13C-labeled root litter addition treatments (to A horizon, to B horizon, and no-addition disturbance control) for each of two harvest time points. We buried root litter in February 2014 and measured loss of 13C in CO2 from the soil surface and in leachate as dissolved organic carbon (DOC) over two growing seasons. At the end of each of the 2014 and 2015 growing seasons we recovered the 13C remaining in the soil. Loss of root litter C occurred almost entirely via heterotrophic respiration, with an estimated < 2% lost as DOC during the initial decay period. The added roots were broken down and incorporated into bulk soil material very quickly; only ~ 30% of added root was visible after 6 months. In the first growing season, decomposition occurred faster in the B than in the A horizon, the latter having greater moisture limitation. Subsequently, there was almost no further decomposition in the B horizon. After two growing seasons, less than 20% of the added root litter C remained in the A or B horizons of all environmental treatments. Heating did not stimulate decomposition, likely because it exacerbated the moisture limitation. However, while plots without plants dried down more slowly than plots with plants, their decomposition rate was not significantly greater, possibly due to the lack of rhizosphere processes such as priming. We conclude that in this Mediterranean grassland ecosystem, soil moisture, which is affected by season, depth, heating, and rhizosphere, plays a dominant role in mediating the effect of those factors on root litter decomposition, which after two seasons did not differ by depth or by treatment.The files in this dataset comprise (1) 13C-labeled litter additions made to each plot, (2) litter recovery after each of 2 growing seasons, (3) periodical soil flux measurements, (4) leachate measurements, and continuously monitored soil (5) temperature and (6) moisture.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2020

Multi-device study of temporal characteristics of magnetohydrodynamic modes initiating disruptions

Disruptions in tokamaks are often preceded by magnetohydrodynamic (MHD) instabilities that can rotate or become locked to the wall. Measurements from magnetic diagnostics in the presence of MHD mode precursors to disruptions can yield potentially valuable input to the plasma control system, with a view to disruption avoidance, prediction and mitigation. This report presents an exploratory analysis of the growth of MHD modes and corresponding time scales on the basis of magnetic measurements in multiple tokamaks. To this end, a database was compiled using disruptive discharges from COMPASS, ASDEX Upgrade, DIII-D and JET, manually classified according to disruption root cause, and characterized by a great diversity of operational conditions and mode dynamics. The typical time during which a mode can be detected using saddle coils and the duration of the locked mode phase in the database both extend over several orders of magnitude, but generally the time scales increase with plasma size. Several additional factors are discussed that can influence these durations, including the disruption root cause. A scaling law for the locked phase duration was estimated, yielding predictions toward ITER of the order of hundreds of milliseconds or even seconds. In addition, a scaling law for the mode amplitude at the disruption onset, proposed earlier by de Vries et al. (2016), is applied to the database, and its predictive capabilities are assessed. Despite significant uncertainty on the predictions from both scaling laws, encouraging trends are observed of the fraction of disruptions that may be detected with sufficient warning time to allow mitigation or even avoidance, based solely on observations of MHD mode dynamics. When combined with similar analysis of measurements from diagnostics that are sensitive to other disruption precursors, our analysis methods and results may contribute to the reliability, robustness and generalization of disruption warning schemes for ITER.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2022

A comprehensive and fair comparison of two neural operators (with practical extensions) based on $\mathrm{FAIR}$ data

Neural operators can learn nonlinear mappings between function spaces and offer a new simulation paradigm for real-time prediction of complex dynamics for realistic diverse applications as well as for system identification in science and engineering. Herein, we investigate the performance of two neural operators, which have shown promising results so far, and we develop new practical extensions that will make them more accurate and robust and importantly more suitable for industrial-complexity applications. The first neural operator, DeepONet, was published in 2019 (Lu et al., 2019), and its original architecture was based on the universal approximation theorem of Chen & Chen (1995). The second one, named Fourier Neural Operator or FNO, was published in 2020, and it is based on parameterizing the integral kernel in the Fourier space. DeepONet is represented by a summation of products of neural networks (NNs), corresponding to the branch NN for the input function and the trunk NN for the output function; both NNs are general architectures, e.g., the branch NN can be replaced with a CNN or a ResNet. According to Kovachki et al. (2021), FNO in its continuous form can be viewed conceptually as a DeepONet with a specific architecture of the branch NN and a trunk NN represented by a trigonometric basis. In order to compare FNO with DeepONet computationally for realistic setups, we develop several extensions of FNO that can deal with complex geometric domains as well as mappings where the input and output function spaces are of different dimensions. We also develop an extended DeepONet with special features that provide inductive bias and accelerate training, and we present a faster implementation of DeepONet with cost comparable to the computational cost of FNO, which is based on the Fast Fourier Transform. Here we consider 16 different benchmarks to demonstrate the relative performance of the two neural operators, including instability wave analysis in hypersonic boundary layers, prediction of the vorticity field of a flapping airfoil, porous media simulations in complex-geometry domains, etc. We follow the guiding principles of FAIR (Findability, Accessibility, Interoperability, and Reusability) for scientific data management and stewardship. The performance of DeepONet and FNO is comparable for relatively simple settings, but for complex geometries the performance of FNO deteriorates greatly. We also compare theoretically the two neural operators and obtain similar error estimates for DeepONet and FNO under the same regularity assumptions.

42 ENGINEERING↗

DOE OSTI2025

Novel Tools for Analyzing Life Cycle Energy Use, Carbon Emissions, and Cost of Additive Manufacturing

Decarbonizing industrial manufacturing is a significant challenge in the effort to limit the impacts of global climate change. Additive manufacturing (AM) is one pathway for reducing the impacts of manufacturing as it creates parts layer-by-layer rather than by removing (i.e., subtracting) material from solid stock as with conventional techniques. This reduces material inputs and generates less waste, which can substantially lower life cycle energy consumption and greenhouse gas emissions. However, AM adoption in the manufacturing sector has been slow, partly due to challenges in making a strong business case compared with more traditional and widely available techniques. This paper highlights the need for the development of simple screening analysis tools to speed the adoption of AM in the manufacturing sector by providing decision-makers easy access to important production life cycle emissions, and cost information. Details on the development of two Microsoft Excel software tools are provided: upgrades to an existing tool on the energy and carbon impacts of AM and a new tool for analyzing the major cost components of AM. A case study applies these two tools to the production of a lightweight aerospace bracket, showing how the tools can be used to estimate the environmental benefits and production costs of AM.

GHG emissions↗

DOE OSTI2021

Quantifying sensitivity in numerical weather prediction-modeled offshore wind speeds through an ensemble modeling approach

A decade of research has shown that numerical weather prediction (NWP)-modeled wind speeds can be highly sensitive to the inputs and setups within the NWP model. For wind resource characterization applications, this sensitivity is often addressed by constructing a range of setups and selecting the one that best validates against observations. However, this approach is not possible in areas that lack high-quality hub height observations, especially offshore wind areas. In such cases, techniques to quantify and disseminate confidence in NWP-modeled wind speeds in the absence of observations are needed. We address this need in the present study and propose best practices for quantifying the spread in NWP-modeled wind speeds. We implement an ensemble approach in which we consider 24 different setups to the Weather Research and Forecasting (WRF) model. We construct the ensemble by considering variations in WRF version, WRF namelist, atmospheric forcing, and sea surface temperature (SST) forcing. Our analysis finds that the standard deviation produces more consistent estimates compared to the interquartile range and tends to be the more conservative estimator for ensemble variability. We further find that model spread increases closer to the surface and on shorter time scales. In conclusion, we explore methods to attribute total ensemble variability to the different ensemble components (e.g., atmospheric forcing and SST product) and find that contributions by components also vary depending on time scale. We anticipate that the methods and results presented in this paper will provide a reasonable foundation for further research into ensemble-based wind resource data sets.

17 WIND ENERGY↗